The objective of this record-linkage investigation is to identify and quantify the risk of cancer following the diagnosis of selected medical conditions and certain surgical procedures. Previous studies of patients with certain medical conditions have revealed significant excesses or decreased risks of cancer. The availability of population-based rosters of patients diagnosed with various medical conditions and/or treated with registry data from the same population provides unique opportunities to conduct cost-efficient record-linkage studies of cancer etiology. Cohorts of patients hospitalized for specific conditions or surgical procedures, or included on rosters or registries of births or congenital disorders have been successfully linked with population-based cancer, mortality, and population registries in Sweden, Denmark, the U.S., and Canada to suggest new hypotheses about cancer etiology, to confirm hypotheses about disease or surgical procedure associations with cancer risk, to demonstrate potentially protective associations for cancer risk, to indicate, albeit indirectly, possible associations of medications with cancer risk, to provide clues as to etiologic mechanisms of carcinogenesis, and to identify populations that may require ongoing medical surveillance because of increased risk of cancer occurrence.